A distress detection system includes a data repository operable to collect sensor data from a monitored system. The distress detection system also includes an analysis system with a processing system operable to access a first parameter of a first system of the monitored system from the data repository and access a second parameter of the first system of the monitored system from the data repository. The processing system is also operable to apply fuzzy reasoning rules to evaluate a combination of the first parameter and the second parameter to determine an in-range interaction with respect to a second system of the monitored system as fuzzy metric data points, classify a component of the second system as being in distress based on comparing the fuzzy metric data points to a limit line, and assert a component distress indicator responsive to classifying the component of the second system as being in distress.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A distress detection system comprising: a data repository operable to collect a plurality of sensor data from a monitored system; and an analysis system comprising a processing system operable to: access a first parameter of a first system of the monitored system from the data repository; access a second parameter of the first system of the monitored system from the data repository; apply a plurality of fuzzy reasoning rules to evaluate a combination of the first parameter and the second parameter to determine an in-range interaction with respect to a second system of the monitored system as a plurality of fuzzy metric data points, wherein the monitored system is a gas turbine engine, the first system is a lubrication system, and the component of the second system is a burner configured to combust fuel in the gas turbine engine; classify a component of the second system as being in distress based on comparing the fuzzy metric data points to a limit line, wherein the combination of the first parameter and the second parameter of the lubrication system indicate that the burner is in distress based on the fuzzy reasoning rules while the first parameter and the second parameter of the lubrication system are in a non-failure condition; and assert a component distress indicator responsive to classifying the component of the second system as being in distress.
2. The distress detection system of claim 1 , wherein the first system is thermally cross-coupled to the second system absent a mechanical linkage, an electrical linkage, and a fluid linkage between the first system and the second system.
3. The distress detection system of claim 1 , wherein the first parameter is a main oil pressure and the second parameter is a main oil temperature.
4. The distress detection system of claim 1 , wherein observations of the first parameter and the second parameter are constrained based on determining that the first parameter and the second parameter are both within a valid operating range.
5. The distress detection system of claim 1 , wherein the first parameter and the second parameter are normalized with respect to a third parameter.
6. The distress detection system of claim 1 , wherein a maintenance action is triggered responsive to the component distress indicator being asserted.
7. The distress detection system of claim 1 , wherein assertion of the component distress indicator is prevented based on detection of a fault condition associated with the first system.
8. The distress detection system of claim 1 , wherein the first parameter and the second parameter are derived from separate sensors of the monitored system.
9. A method of distress detection in a monitored system, the method comprising: accessing, by an analysis system, a first parameter of a first system of the monitored system; accessing, by the analysis system, a second parameter of the first system of the monitored system; applying a plurality of fuzzy reasoning rules to evaluate a combination of the first parameter and the second parameter to determine an in-range interaction with respect to a second system of the monitored system as a plurality of fuzzy metric data points, wherein the monitored system is a gas turbine engine, the first system is a lubrication system, and the component of the second system is a burner configured to combust fuel in the gas turbine engine; classifying a component of the second system as being in distress based on comparing the fuzzy metric data points to a limit line, wherein the combination of the first parameter and the second parameter of the lubrication system indicate that the burner is in distress based on the fuzzy reasoning rules while the first parameter and the second parameter of the lubrication system are in a non-failure condition; and asserting a component distress indicator responsive to classifying the component of the second system as being in distress.
10. The method of claim 9 , wherein the first system is thermally cross-coupled to the second system absent a mechanical linkage, an electrical linkage, and a fluid linkage between the first system and the second system.
11. The method of claim 9 , wherein the first parameter is a main oil pressure and the second parameter is a main oil temperature.
12. The method of claim 9 , wherein observations of the first parameter and the second parameter are constrained based on determining that the first parameter and the second parameter are both within a valid operating range.
13. The method of claim 9 , wherein the first parameter and the second parameter are normalized with respect to a third parameter.
14. The method of claim 9 , wherein a maintenance action is triggered responsive to the component distress indicator being asserted.
15. The method of claim 9 , wherein assertion of the component distress indicator is prevented based on detection of a fault condition associated with the first system.
16. The method of claim 9 , wherein the first parameter and the second parameter are derived from separate sensors of the monitored system.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 8, 2018
November 10, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.